NSG 410 Stratford University Pain Associated with Cancer Paper ARTICLE: A Randomized, Clinical Trial of Education or Motivational-Interviewing–Based Coachi

NSG 410 Stratford University Pain Associated with Cancer Paper ARTICLE: A Randomized, Clinical Trial of Education or Motivational-Interviewing–Based Coaching Compared to Usual Care to Improve Cancer Pain Management . By Mary Laudon Thomas, RN, MS, AOCN®, Janette E. Elliott, RN-BC, MS, AOCN®, APA FORMAL( TITLE, ABSTRACT, INTRODUCTION’, BODY, WELL SOURCE/REFERENCE.)… Downloaded on 10 13 2016. Single-user license only. Copyright 2016 by the Oncology Nursing Society. For permission to post online, reprint, adapt, or reuse, please email pubpermissions@ons.org
© Oncology Nursing Society. Unauthorized reproduction,
in part or in whole, is strictly prohibited. For permission
to photocopy, post online, reprint, adapt, or otherwise reuse any
or all content from this article, e-mail pubpermissions@ons.org.
To purchase high-quality reprints, e-mail reprints@ons.org.
Journal Club Article
See page 49 for suggested questions to begin
discussion in your journal club.
A Randomized, Clinical Trial of Education
or Motivational-Interviewing–Based Coaching Compared
to Usual Care to Improve Cancer Pain Management
Mary Laudon Thomas, RN, MS, AOCN®, Janette E. Elliott, RN-BC, MS, AOCN®,
Stephen M. Rao, PhD, Kathleen F. Fahey, RN, MS, CNS, Steven M. Paul, PhD,
and Christine Miaskowski, RN, PhD, FAAN
D
espite important advances in its management, cancer pain remains a significant clinical problem (Apolone
et al., 2009; McGuire, 2004; van den
Beuken-van Everdingen et al., 2007). In
a meta-analysis, cancer pain was found in 64% of patients with metastatic disease, 59% of patients receiving
antineoplastic therapy, and 33% of patients who had
received curative cancer treatment (van den Beukenvan Everdingen et al., 2007). Cancer pain also has a
negative effect on patients’ functional status (Ferreira
et al., 2008; Holen, Lydersen, Klepstad, Loge, & Kassa,
2008; Vallerand, Templin, Sasenau, & Riley-Doucet,
2007) and is associated with psychological distress
(Cohen et al., 2003; Vallerand, Hasenau, Templin, &
Collins-Bohler, 2005). The effect of cancer pain on an
individual’s quality of life (QOL) can be significant
and extend beyond disturbances in mood and physical
function (Burckhardt & Jones, 2005; Dahl, 2004; Fortner
et al., 2003).
Although advances in pain management can reduce cancer pain for a significant number of patients,
numerous clinician, healthcare system, and societal
barriers (e.g., knowledge deficits, reimbursement and
regulatory constraints, religious or cultural views)
contribute to ineffective pain management (Brockopp
et al., 1998; Dahl, 2004; Hill, 1993; Sun et al., 2007). Attitudinal barriers held by patients can be a substantive
factor in the inadequate treatment of cancer pain (Anderson et al., 2002; Ward et al., 2008). Those attitudinal
barriers need to be addressed if cancer pain management is to be improved (Fahey et al., 2008).
In a meta-analysis of the benefits of patient-based
psychoeducational interventions for cancer pain management, Bennett, Bagnall, and Closs (2009) concluded
Oncology Nursing Forum • Vol. 39, No. 1, January 2012
Purpose/Objectives: To test the effectiveness of two
interventions compared to usual care in decreasing attitudinal barriers to cancer pain management, decreasing
pain intensity, and improving functional status and quality
of life (QOL).
Design: Randomized clinical trial.
Setting: Six outpatient oncology clinics (three Veterans Affairs
[VA] facilities, one county hospital, and one community-based
practice in California, and one VA clinic in New Jersey)
Sample: 318 adults with various types of cancer-related pain.
Methods: Patients were randomly assigned to one of three
groups: control, standardized education, or coaching.
Patients in the education and coaching groups viewed a
video and received a pamphlet on managing cancer pain.
In addition, patients in the coaching group participated in
four telephone sessions with an advanced practice nurse
interventionist using motivational interviewing techniques
to decrease attitudinal barriers to cancer pain management.
Questionnaires were completed at baseline and six weeks
after the final telephone calls. Analysis of covariance was
used to evaluate for differences in study outcomes among
the three groups.
Main Research Variables: Pain intensity, pain relief, pain
interference, attitudinal barriers, functional status, and QOL.
Findings: Attitudinal barrier scores did not change over time
among groups. Patients randomized to the coaching group
reported significant improvement in their ratings of painrelated interference with function, as well as general health,
vitality, and mental health.
Conclusions: Although additional evaluation is needed,
coaching may be a useful strategy to help patients decrease
attitudinal barriers toward cancer pain management and to
better manage their cancer pain.
Implications for Nursing: By using motivational interviewing techniques, advanced practice oncology nurses can help
patients develop an appropriate plan of care to decrease pain
and other symptoms.
39
that, compared to usual care, educational interventions
improved knowledge and attitudes and reduced average and worst pain intensity scores. However, those
interventions had no effect on medication adherence or
in reducing pain’s level of interference with daily activities. Bennett et al. (2009) suggested that additional trials
are warranted to test different approaches to cancer pain
education and to clarify the exact relationships between
education and improved patient outcomes.
Many psychoeducational intervention studies were
conducted in the hospital setting (Chang, Chang, Chiou,
Tsou, & Lin, 2002; de Wit et al., 2001; Jahn et al., 2010) or
in patients’ homes (Given et al., 2002; Miaskowski et al.,
2004), which limited the generalizability of the findings
to the outpatient clinic setting. In addition, although
they achieved a positive outcome, many of the studies
were labor-intensive, which also limited their ability to
be implemented in a busy oncology clinic (Given et al.,
2002; Miaskowski et al., 2004). Unfortunately, studies
using less labor-intensive interventions were not as
successful in decreasing cancer pain (Anderson et al.,
2002; Oliver, Kravitz, Kaplan, & Meyers, 2001; Syrjala
et al., 2008).
Coaching is a useful strategy to improve cancer pain
management (Kalauokalani, Franks, Oliver, Meyers, &
Kravitz, 2007; Miaskowski et al., 2004). Incorporating
principles of motivational interviewing into a coaching
intervention affords a unique method of exploring personal attitudes, behaviors, and beliefs that can interfere
with effective cancer pain management (Fahey et al.,
2008; Prochaska & DiClemente, 1984).
Change theory, specifically the Transtheoretical Model
(Prochaska & DiClemente, 1984), is a useful conceptual framework for coaching. In this model, behavioral
change is a function of a person’s state of readiness or
motivation to modify a particular behavior. Motivational interviewing is a nonauthoritarian counseling
technique that can assist patients in recognizing and
resolving ambivalence about making constructive behavioral changes. It matches the patients’ readiness to
change and can motivate the patient to move through
the stages of the Transtheoretical Model: precontemplation (unaware of need for change), contemplation
(thinking about change), preparation (actively considering change), action (engaging in changing behavior),
and maintenance (maintaining a changed behavior)
(Fahey et al., 2008; Prochaska & DiClemente, 1984).
Given the limitations of previous intervention studies,
additional research is warranted using approaches that
can be implemented in the outpatient setting. Therefore,
the purposes of this randomized clinical trial were to test
the effectiveness of two interventions compared to usual
care in decreasing attitudinal barriers to cancer pain
management, decreasing pain intensity, and improving pain relief, functional status, and QOL. The authors
hypothesized that the motivational-interviewing–based
40
coaching group would demonstrate greater benefit (i.e.,
decreasing attitudinal barriers; decreasing pain intensity; and improving pain relief, functional status, and
QOL) than either the conventional education or usual
care groups.
Methods
Sample and Settings
A convenience sample was obtained by recruiting
patients from six outpatient oncology clinics (three
Veterans Affairs [VA] facilities, one county hospital,
and one community-based practice in California, and
one VA clinic in New Jersey). Patients were eligible to
participate if they were able to read and understand the
English language, had access to a telephone, had a life
expectancy longer than six months, and had an average
pain intensity score of 2 or higher as measured on a 0–10
scale, with higher scores indicating more pain. Patients
were excluded if they had a concurrent cognitive or
psychiatric condition or substance abuse problem that
would prevent adherence to the protocol, had severe
pain unrelated to their cancer, or resided in a setting
where the patient could not self-administer pain medication (e.g., nursing home, board and care facility). The
study was approved by the institutional review board
and research committee at each of the sites. To test the
interaction of time (change in scores from pre- to poststudy) by assignment to the three treatment groups
(i.e., control, education, or coaching), a sample size of
240 was needed to detect a medium effect (f = 0.25; h2 =
6% of explained variance). As shown in Figure 1, of the
1,911 patients who were screened, 406 were eligible to
participate, 322 provided written informed consent, and
289 completed baseline assessments after being randomized to one of three groups.
Procedures
Prior to beginning participant recruitment, all research team members were trained extensively so
that the procedures for enrollment, data collection,
and interventions were standardized across all clinic
sites. Research associates (RNs or psychology interns)
were trained in procedures for evaluating potential
participants, approaching them, obtaining consent to
participate, and administering the instruments and
videotapes. Importantly, the research associates were
trained in providing attention-control telephone calls.
The nurse interventionist was trained extensively in
motivational interviewing and change theory by a
cognitive behavioral psychologist and then in procedures related to the specific coaching protocol. Details
of this training are described in Fahey et al. (2008).
Monthly team meetings were held throughout the
study to ensure procedural fidelity was maintained.
Vol. 39, No. 1, January 2012 • Oncology Nursing Forum
Patients and clinicians at the
study sites were blinded to the
patient’s group assignment.
At the time of enrollment, pa• Did not meet inclusion criteria (n = 1,505)
tients completed a demographic
• Declined to participate (n = 84)
questionnaire, the Karnofsky
Performance Status (KPS) scale
(Karnofsky & Burchenal, 1949),
Stratify (N = 322)a
the Brief Pain Inventory (Daut,
• Treatment (chemotherapy, radiation therapy,
Cleeland, & Flanery, 1983), the
or none)
Barriers Questionnaire (BQ)
• Pain (low, medium, or high)
(Ward et al., 1993), the 36-Item
Short Form Health Survey (SF36®) (Ware & Sherbourne, 1992),
Randomize into groups (N = 318)a
and the Functional Assessment
of Cancer Therapy–General
(FACT-G) (Cella et al., 1993).
Control (n = 109)
Education (n = 103)
Coaching (n = 105)
The patients’ medical records
were reviewed for disease and
treatment information.
Completed T1 (n = 104)
Completed T1 (n = 94)
Completed T1 (n = 91)
Patients in the usual care group
viewed a video on cancer (American Cancer Society, 1994). PaCompleted coaching
tients assigned to the education
intervention (n = 74)
group viewed a video on managing cancer pain that focused on
overcoming attitudinal barriers
Completed T2 (n = 88)
Completed T2 (n = 75)
Completed T2 (n = 64)
(Syrjala, Abrams, Du Pen, Niles,
a
Four patients withdrew before randomization, and one was lost to follow-up before completing T1.
& Rupert, 1995) and received the
Note. Reasons for lack of completion included being too ill, withdrawing, fatigue, being lost to
Agency for Health Care Policy
follow-up, death, ineligibility, prolonged hospitalization, protocol violation, or other.
and Research (1994) pamphlet
Figure 1. Trial Participation at Baseline (T1) and Six Months (T2)
entitled, Managing Cancer Pain,
Consumer Version, Clinical Practice
Guideline Number 9. To simulate
the time constraints in many oncology outpatient clinPatients were identified by clinic staff and screened
ics, no reinforcement of the material was provided
for eligibility by the research associate, who then apunless the patient sought additional information or
proached eligible patients, explained the study, and
asked questions of the clinic staff. Patients assigned
obtained written informed consent. Patients were stratito the coaching group received the same intervention
fied based on pain intensity (i.e., low, medium, or high)
as those assigned to the education group. In addition,
and cancer treatment (i.e., chemotherapy or radiation
they participated in four 30-minute telephone sessions
therapy) to control for the confounding variables of pain
that explored beliefs about pain, use of analgesics and
intensity and the effects of cancer treatment. Stratifying
nonpharmacologic pain management strategies, and
by pain intensity accounts for the curvilinear relationcommunication about pain management. Those four
ship between pain severity and functional status (e.g.,
calls were conducted about every other week over a
changes in pain intensity at the upper levels of the scale
six-week time period by the nurse interventionist, a
have a different effect on functional status compared to
clinical nurse specialist trained in motivational interchanges at the lower levels of the scale). Stratification
viewing techniques. For a detailed description of the
by cancer therapy was used to control for the effect of
coaching intervention, see Fahey et al. (2008). Patients
treatment in either decreasing pain from shrinking the
assigned to the usual care and education groups also
tumor or increasing pain because of toxicity of treatreceived four telephone calls (about every other week
ment. Patients at each clinic site then were randomized
over a six-week time period) from the research asbased on the stratification criteria using permuted
sistance for attention-control purposes. Six weeks
blocks with variable sizes into one of three groups: usual
after the final telephone call (i.e., 12 weeks postrancare (control), education, or coaching. This method of
domization), all patients completed the same quesrandomization was used to ensure balance across the
tionnaires that were done at enrollment. Participants
treatment groups within each stratification cell.
Assessed for eligibility (N = 1,911)
Oncology Nursing Forum • Vol. 39, No. 1, January 2012
41
received a $25 gift certificate after completing each set
of questionnaires.
The BQ has demonstrated adequate validity and reliability (Ward et al., 1993; Ward & Gatwood, 1994).
Pain was assessed with the Brief Pain Inventory, a
self-report instrument designed to assess the intensity
and quality of pain, the extent to which pain relief was
obtained, and the extent to which pain interferes with
function (Daut et al., 1983). Severity and interference
are rated on a numeric score from 0 (does not interfere)
to 10 (completely interferes). A mean interference score
was calculated (Serlin, Mendoza, Nakamura, & Cleeland, 1995), with higher scores reflecting greater pain
intensity and greater interference with function.
Functional status was measured with the SF-36
(Ware & Sherbourne, 1992). Eight health concepts
Instruments
Attitudinal barriers were assessed with the BQ (Ward
et al., 1993; Ward & Gatwood, 1994), a 27-item instrument
that measures eight barriers to cancer pain management
(concern about side effects, concern about tolerance, fear
of addiction, fatalism, fear of disease progression, desire
to be a good patient, fear of injections, and concern about
distracting the physician from curing disease). Each item
is rated on a scale from 0 (not at all agree) to 5 (agree very
much). Mean subscale and total scores were calculated
for the BQ, with higher scores reflecting stronger barriers.
Table 1. Demographic and Clinical Characteristics by Study Group
Characteristic
Age (years)
Education (years)
Time since diagnosis (months)
Karnofsky Performance Status scorec
Characteristic
Gender
Male
Female
Ethnicity
African American
Caucasian
Latino
Other
Marital status
Married or partnered
Widowed, divorced, or separated
Never married
Living arrangements
Alone
With family or friends
Other
Employment
Full- or part-time
Disability, leave of absence, or retired
Unemployed
Other
Cancer diagnosis
Breast
Colon
Head and neck
Lung
Myeloma
Prostate
Other (mixed types)
Control
(N = 88)a
Education
(N = 75)b
Coaching
(N = 64)
—
—
—
X
SD
X
SD
X
SD
Statistics
58.7
13.8
31.9
76.6
11.5
2.7
52.7
12.5
62.5
12.8
37.5
72.3
11.2
2.6
45
12.7
61.8
13.1
30
77.6
11.3
3.2
42.5
13.2
F(2, 223) = 2.54, p = 0.08
F(2, 222) = 2.57, p = 0.08
F(2, 222) = 0.48, p = 0.62
F(2, 222) = 3.53, p = 0.03*
n
%
n
%
n
%
Statistics
79
9
90
10
71
4
95
5
54
10
84
16
21
48
6
11
24
56
7
13
15
44
8
7
20
60
11
10
7
44
7
6
11
69
11
9
40
33
15
46
38
17
37
23
14
50
31
19
33
27
4
52
42
6
23
55
10
26
63
11
12
57
6
16
76
8
15
47
2
23
73
3
10
54
18
4
12
63
21
5
4
57
11
2
5
77
15
3
5
48
10
1
8
75
16
2
5
6
12
21
6
12
26
6
7
14
24
7
14
30
3
2
7
14
5
16
28
4
3
9
19
7
21
37
8
4
6
9
6
11
20
13
6
9
14
9
17
31
c2 = 4; p = 0.13
c2 = 13.4, p = 0.65
c2 = 8.3, p = 0.61
c2 = 6.4, p = 0.38
c2 = 10.1, p = 0.61
c2 = 45.7, p = 0.72
* Education < coaching, p < 0.05 a Because patients could refuse to complete items, N = 86 for ethnicity and employment. b Because patients could refuse to complete items, N = 74 for ethnicity, marital status, and employment. c Scores indicate functional status on a 0–100 scale, with higher scores reflecting higher function. Note. Because of rounding, not all percentages total 100. 42 Vol. 39, No. 1, January 2012 • Oncology Nursing Forum Of the 289 patients who enrolled, 227 completed the end-of-study evaluation. The length of time from cancer diagnosis to study enrollment averaged 30–38 months. The most common cancer types were lung, prostate, and head and neck. Most patients were men and middleaged, and about half of the sample was married or partnered. No differences were found among the three groups on any demographic or clinical characteristic except KPS score. Patients in the education group reported significantly lower KPS scores than patients in the coaching group (p = 0.03) (see Table 1). Instrument Scores Barrier Questionnaire: Barrier subscale scores were modest in all three groups, with concerns about addiction and disease progression rated higher than those related to fatalism or the need to be a “good patient” (data not shown). However, after controlling for each of the BQ scores at baseline, no differences were found among the three groups in any of the subscale or total BQ scores. Pain intensity, interference, and relief: After controlling for average pain at baseline, no differences were found among the three groups in average pain intensity scores at the end of the study (p = 0.08) (see Figure 2). Similarly, nonsignificant scores were found among the three groups in worst pain intensity scores (data not shown). However, significant differences were found among the three groups in mean pain interference scores at the end of the study (p = 0.01) (see Figure 3). Post-hoc Average Pain Intensity 7 6 5 s 4 s 3 2 1 0 Baseline Control End of Study s s Oncology Nursing Forum • Vol. 39, No. 1, January 2012 Sample s Differences in demographic and clinical characteristics among the three groups were evaluated using analyses of variance and chi-square tests. Analyses of covariance were performed to evaluate for differences in scores on average and worst pain intensity, pain relief, mean pain interference, the BQ, the SF-36, and the... Purchase answer to see full attachment